Ever since the s eminal paper by Friedman (1968) and Phelps (1967), then atural rate hypothesis has occupied a pivotal position in macroeconomics. This hypothesis states that onl y unexpected changes in the rate of inflation affect real variables su ch as unemplo yment and real output. In its rational expectations version, developed by Lucas (1972,1973), the hypothesis implies that policy makers should ignore any temporary tradeo ffs and strive only for low inflation.

Other economists, such as Ball (1997) and Romer and Romer (1994) argue that fo ravariety of reasons (the nature of labor markets, menu costs, hysteresis, etc.) tight monetary policies, aimed at reducing inflation, can also have a lasting effect on employment and output.

There have been numerous empirical tests of the natural ratehypothesis 1 , although a problem common to virtually all of them is the lack of a reliable measure of inflationary expectations. As a result of this problem, one is alwa ys f aced with the difficulty of disentan gling the r esults of a jointhypothesis - that expected inflation does not have any real effects and that inflationary expectations are formed in one way or another.

The pur pose of this paper is to test the naturality hypothesis by using market extracted inflationary expectations from I sraeli data. As is ar gued b elow, the se e xpectations e nable one to conduct a better test of the natur al rate hypothesis than the usual alternatives.

These expectations are derived from the yields of nominal and CPI linked bonds o f identical maturities. Such bonds are traded regularly on the Tel-Aviv stock exchange. The paper is or ganized as follows: in section 2 we discuss the advantages of using the market extracted inflationary expectations. Section 3 describes the data. Section 4 desc ribes the econometric approach and presents the empirical results. Section 5 provides a brief conclusion.

The full article in PDF file